GRP proteins (glucose-regulated proteins) are molecular chaperones localized to endoplasmic reticulum (ER). They are known as members of a protein family that is induced in response to various intrinsic or extrinsic ER stresses, such as glucose starvation, or accumulation of misfolded proteins in ER (Non-patent document 1).
GRP78 is one of GRP proteins with a molecular weight of 78 kDa and is also well known as BiP (immunoglobulin binding protein). Overexpression, or antisense approaches directly showed that GRP78 is involved in protective role from cell death caused by ER stress (Non-patent document 1).
Solid tumor cells in vivo are constantly exposed in ER stress, including glucose deprivation, hypoxia and low pH, due to a feature of the tumor microenvironment. As if to support this understanding, increased expression of GRP78 protein has been confirmed in a variety of cancer cell lines or clinical cancer specimens, correlating with malignancy (Non-patent documents 2 to 5). Further, it has been demonstrated that an overexpression of GRP78 protein is involved in the acquisition of resistance to therapeutic treatment of anticancer drugs generated by topoisomerase inhibitory activity or antiangiogenesis agents (Non-patent documents 6 and 7). In a clinical study, a group of breast cancer patients with enhanced expression of GRP78 were demonstrated to be less responsive to adriamycin-based chemotherapy than a group with lower expression of GRP78 (Non-patent document 8).
These reports suggest that up-regulation of GRP78 expression in tumor is associated with the mechanisms of survival, malignant transformation, resistance to anticancer agents (Non-patent document 1).
GRP78 is a molecular chaperone localized in ER, whereas translocation of this molecule on the cancer cell membrane has been reported. Furthermore, the possibility of application for cancer therapy by targeting the surface-located GRP78 has been indicated by several groups, with entirely different approach.
When the rabdomisarcoma cell line TE 671/RD was treated with thapsigargin (Tg), the cell membrane was slightly stained with an anti-GRP78 antibody, as confirmed by FACS analysis, thus demonstrating cell membrane localization of the GRP78 (Non-patent document 9).
It should be noted here that this report addresses a transient event during the induction of cell death by Tg treatment and that it does not show data for a persistent change in GRP78 localization in cancer cells. In addition, since the antibody used in the report is a commercially available goat-derived polyclonal antibody, its whose epitope is unknown.
Later, another research group reported that the two GRP78 binding peptides (WIFPWIQL (SEQ ID NO: 103) and WDLAWMFRLPVG (SEQ ID NO: 104)) which acquired by phage binding assays could bind to the cell surface of the prostate cancer cell line DU145 and being internalized into the cells (Non-patent document 10).
In addition, those GRP78 binding peptides fused to the cell death-inducing motif (KLAKLAK)2 (SEQ ID NO: 105) (Non-patent document 11) have shown not only the induction of cell death on DU145 cells in vitro but also antitumor effects in an experiment on mouse transplant models (Non-patent document 10).
Another research group reported that the surface-located GRP78 protein on vascular endothelial cells serves as a receptor for angiogenesis inhibitor Kringle 5 (K5) (Non-patent document 16). They further demonstrated that interaction of GRP78 with a recombinant K5 induces not only an inhibition of angiogenesis but also cell death on various cancer cell lines cultured under hypoxia (Non-patent document 16).
Thus, the series of experiments described above have shown that peptides that bind to the surface-located GRP78 on cancer cells or vascular endothelial cells might be a useful tool as antitumor agents. However, it would be difficult to apply them in the development for clinical application since the site of the surface-located GRP78 protein recognized by these peptides is not known in the art.
Aside from those findings, two entirely unrelated groups relied upon different approaches to report localization of GRP78 on the cell membrane.
One group showed that the receptor of the activated forms of α2-macroglobulin (α2M*), which functioned as a growth factor in a prostate cancer cell line (1-LN) (Non-patent document 12), was GRP78 (Non-patent document 13). The discovery added a new finding that the GRP78 protein long considered to be an ER protein also functions as a receptor of the growth factor on the cell membrane.
The other group studied that polyclonal antibody in the serum from prostate cancer patient recognize the peptide sequence “CNVSDKSC” (SEQ ID NO: 106) (i.e., anti-CNVSDKSC (SEQ ID NO: 106) antibody); they then identified that a target protein recognized by this antibody was the GRP78 protein (Non-patent document 14). Although the anti-CNVSDKSC (SEQ ID NO: 106) antibody bound to cell surface, there were no matched or similar sequence “CNVSDKSC” (SEQ ID NO: 106) within GRP78 sequence, suggesting that primary amino acid sequence “CNVSDKSC” (SEQ ID NO: 106) contains a tertiary structural motief mimicking an epitope in surface-located GRP78. However, it was not identified the site of GRP78 which were recognized by this antibody.
Later in time, another group performed a tertiary structural analysis of the peptide “CNVSDKSC” (SEQ ID NO: 106) and identified the GRP78 primary amino acid sequence “LIGRTWNDPSVQQDIKFL” (SEQ ID NO: 107) located at Leu98-Leu115 which forms the similar tertiary structure serving as a putative binding site. They then prepared a rabbit polyclonal antibody against this sequence and eventually confirmed that the antibody could stain the cell surface of cancer cells, i.e., the prostate cancer cell lines 1-LN and DU145 as well as the melanoma cell line DM413. It was also confirmed that the antibody, when added to the prostate cancer cell lines, had an ability to increase the intracellular calcium concentration, induce cell proliferation, and protect the cell from apoptosis induced by TNF-.alpha., as observed upon addition of α2M* (Non-patent document 15). Since the antibody against GRP78 thusly mimicked the ligand activity of α2M*, it was revealed that the region Leu98-Leu115 of GRP78 was an α2M*binding sequence (Non-patent document 15).
This report validated that GRP78 is localized to cell surface in prostate cancer and it was further revealed that Leu98-Leu115 of GRP78 (LIGRTWNDPSVQQDIKFL) (SEQ ID NO: 107) was exposed extracellularly as an α2M* binding sequence.
Further, from another approach, it was reported that antibodies against the 98-115 region of GRP78 stained the cell surface of cancer cells, thus revealing that this region would be capable of serving as an extracellular epitope of GRP78.
Thus it was found that the GRP78 protein is highly expressed in many cancer types, with a localization change on the cell membrane. However it has been difficult to develop a new therapeutic antibody targeted to the surface-located GRP78 on the basis of findings as below. First, the site of the GRP78 exposed on the cell surface, which were recognized by the above-described GRP78 binding peptides is not clarified, rendering it impossible to prepare antibodies that provide similar effects to the peptides; indeed, no monoclonal antibody which functionally mimic those peptides exists. Second, the antibody which recognizes the 98-115 region of GRP78 can bind to the surface-located GRP78 of cancer cells but, at the same time, it mimics the α2M* growth promoting action, so this antibody cannot be expected to display an antitumor activity.
Hence, it has been considered difficult to exert antitumor activity by means of the GRP78 binding antibody.
Non-patent document 1: Lee A S. Trends Biochem Sci. 2001, 26, 504-10
Non-patent document 2: Patierno, et al. 1998, Cancer Res. 47, 6220-24
Non-patent document 3: Bini et al. 1997, Electrophoresis. 18, 2832-41
Non-patent document 4: Gazit et al. 1999, Breast Cancer Res. Treat. 54, 135-46
Non-patent document 5: Fernandez et al. 2000, Breast Cancer Res. Treat. 59, 15-26
Non-patent document 6: Reddy et al, J. Bio. Chem. 2003, 278, 20915-24
Non-patent document 7: Dong et al, 2005, Cancer Res. 65, 5785-5791
Non-patent document 8: Lee et al. 2006, Cancer Res. 66, 7849-7853
Non-patent document 9: Delpino et al. 1998, Molecular Membrane Biology, 15, 21-26
Non-patent document 10: Arap et al. 2004, CANCER CELL. 6, 275-284
Non-patent document 11: Javadpour et al. 1996, J. Med. Chem. 39, 3107-3113
Non-patent document 12: Asplin et al. 2000, Archives of Biochemistry and Biophysics. 383, 135-141
Non-patent document 13: Misra et al. 2002, J. Biol. Chem. 277, 42082-42087
Non-patent document 14: Mintz et al. 2003, Nat. Biotech. 21, 57-63
Non-patent document 15: Gonzalez-Gronow et al. 2006, Cancer Res. 66, 11424-11431
Non-patent document 16: Davidoson et al., 2005, Cancer Res. 65, 4663-4672